Article Review: Designing Ethical AI Systems

Oberoi, S. S., Singh, A. N., & Chakraborty, D. (2025). Designing ethical AI systems: An exploration of deontological, virtue, utilitarian, and rights-based ethical frameworks. Journal of Global Information Management (JGIM), 33(1), 1-25. https://doi.org/10.4018/JGIM.388742

Introduction and Purpose of the Article

The article Designing ethical AI systems: An exploration of deontological, virtue, utilitarian, and rights-based ethical frameworks addresses a timely and significant issue: the absence of structured and theory-driven ethical guidelines for the design and deployment of artificial intelligence (AI) systems. As AI technologies become increasingly embedded across sectors such as healthcare, education, and business, ethical concerns related to transparency, bias, accountability, and human rights have intensified. Oberoi, Singh, and Chakraborty seek to address this gap by proposing a conceptual framework that integrates four classical ethical theories-deontology, virtue ethics, utilitarianism, and rights-based ethics into the AI system design process.

The primary objective of the study is to examine how these ethical frameworks influence ethical AI implementation and to assess how openness and ethical relativism mediate the relationship between AI system design and ethical outcomes. Oberoi et al. position the study as one of the first empirical investigations to directly link traditional ethical theories with AI system design, thereby contributing to both ethics scholarship and the information systems literature.

Theoretical Framework and Literature Review

The article is grounded in moral philosophy and business ethics, providing a robust theoretical foundation. Oberoi et al. describe deontology as a duty-based framework emphasizing rules, obligations, and harm prevention, making it particularly relevant for addressing unethical AI behaviors such as deception or discrimination. Virtue ethics is presented as focusing on moral character and organizational values, including honesty, compassion, and integrity in AI development. Utilitarianism prioritizes maximizing overall societal benefit, while rights-based ethics emphasizes the protection of fundamental human rights such as privacy, autonomy, and freedom from bias.

The literature review is comprehensive and interdisciplinary, drawing on philosophy, management, and information systems research. Oberoi, Singh, and Chakraborty clearly identify a gap in existing scholarship by noting that, despite extensive discussions of AI ethics, few studies offer an integrated, theory-driven framework that has been empirically validated. This identified gap provides a strong justification for the study.

Research Design and Methodology

Oberoi et al. employ a mixed-methods research design consisting of an exploratory qualitative phase followed by a confirmatory quantitative phase. In the first phase, in-depth interviews were conducted with 31 information technology professionals using grounded theory to identify key ethical constructs relevant to AI design. This qualitative approach strengthens the conceptual validity of the proposed framework by grounding it in practitioners’ real-world experiences.

In the second phase, Oberoi, Singh, and Chakraborty collected survey data from 493 AI users across India and analyzed the data using Partial Least Squares Structural Equation Modeling (PLS-SEM). Methodological rigor is demonstrated through reliability and validity testing, with Cronbach’s alpha, composite reliability, and average variance extracted (AVE) exceeding accepted thresholds. While India is an appropriate research context given its rapid AI adoption and expanding digital infrastructure, the geographic focus may limit broader generalizability.

Key Findings and Discussion

The findings indicate that all four ethical frameworks, deontology, virtue ethics, utilitarianism, and rights-based ethics, have a statistically significant positive impact on ethical AI system development. Oberoi et al. report that deontology and rights-based ethics exhibit particularly strong effects, underscoring the importance of rule adherence, duty, and the protection of human rights in ethical AI design.

The study further demonstrates that embedding ethical theories into AI system design significantly contributes to ethical outcomes. Oberoi et al. find that openness positively moderates this relationship, suggesting that transparency, stakeholder engagement, and clear communication enhance ethical AI development. In contrast, ethical relativism does not show a significant moderating effect, indicating the necessity of universal baseline ethical principles rather than context-dependent moral standards. The discussion effectively links empirical findings to existing theory and prior research, reinforcing the argument that ethics must be integrated during the design stage rather than treated as an afterthought.

Strengths of the Article

A major strength of the article lies in Oberoi, Singh, and Chakraborty’s integration of classical ethical theories with contemporary AI challenges. The mixed-methods approach enhances both depth and empirical robustness, while the validation of the conceptual model strengthens the study’s credibility. The identification of openness as a key moderating factor provides practical guidance for organizations seeking to operationalize ethical AI principles. The article is well organized, logically structured, and supported by extensive scholarly references, making it valuable to researchers, policymakers, and practitioners.

Limitations and Directions for Future Research

Despite its contributions, the study has several limitations. Oberoi et al. acknowledge that reliance on self-reported survey data introduces the potential for response bias. Additionally, the exclusive focus on India may limit the applicability of the findings across different cultural, legal, and regulatory environments. Future research could replicate the model in other geographic contexts or industries and explore additional ethical frameworks, such as care ethics or justice-based ethics, to further enrich ethical AI scholarship.

Conclusion

Overall, the article makes a timely and substantive contribution to the evolving discourse on ethical artificial intelligence. By grounding AI design in established ethical theories and validating the framework through empirical analysis, Oberoi, Singh, and Chakraborty advance both theoretical understanding and practical application. The study convincingly demonstrates that ethical AI requires deliberate design choices informed by moral principles, transparency, and stakeholder engagement. As AI continues to shape societal structures, this article offers a valuable roadmap for developing systems that are not only intelligent but also ethically responsible.